Applying Data Mining to Explore Students' Self-Regulation in Learning Contexts

Abstract

The theory of self-regulation, which includes cognitive, motivational, and behavioral dimensions, is useful for understanding the relationships between motivations, learning strategies, and the learner in the context of learning. The critical attributes, such as planning, metacognitive monitoring, and self-reflection, provide valuable information to clarify why some students perform better than others. In order to extract significant attributes for successful learners, the supervised learning method is applied to 131 students of Tunghai University in Taiwan. The mining results indicate that successful students normally spend time on the unclear concepts and put more emphasis on difficult learning materials during the course of learning. The findings provide further information for understanding how students deploy their self-regulatory behaviors in the learning contexts.